Swarm Intelligence and Deep Evolution: Evolutionary Approach to Artificial Intelligence

✍ Scribed by Hitoshi Iba

• Publisher: CRC Press
• Year: 2022
• Tongue: English
• Leaves: 288
• Category: Library

✦ Synopsis

The book provides theoretical and practical knowledge about swarm intelligence and evolutionary computation. It describes the emerging trends in deep learning that involve the integration of swarm intelligence and evolutionary computation with deep learning, i.e., deep neuroevolution and deep swarms. The study reviews the research on network structures and hyperparameters in deep learning, and attracting attention as a new trend in AI. A part of the coverage of the book is based on the results of practical examples as well as various real-world applications. The future of AI, based on the ideas of swarm intelligence and evolution is also covered.

The book is an introductory work for researchers. Approaches to the realization of AI and the emergence of intelligence are explained, with emphasis on evolution and learning. It is designed for beginners who do not have any knowledge of algorithms or biology, and explains the basics of neural networks and deep learning in an easy-to-understand manner. As a practical exercise in neuroevolution, the book shows how to learn to drive a racing car and a helicopter using MindRender. MindRender is an AI educational software that allows the readers to create and play with VR programs, and provides a variety of examples so that the readers will be able to create and understand AI.

✦ Table of Contents

Cover
Title Page
Copyright Page
Preface
Table of Contents
1. AI: Past and Present
1.1 AI and its History
1.2 Pareto-efficiency and Human Intelligence
2. Evolutionary Theories for AI
2.1 What is Evolution?
2.2 Neutral Molecular Evolution
2.2.1 Moran Process
2.2.2 Genetic Drift and Fixation Probability
2.2.3 Evolution Speed
2.2.4 Neutral Theory
2.2.5 Neutral Evolution by Simulation
2.2.6 Baldwinian Evolution
2.3 Introns and Selfish Genes
2.3.1 Basics of DNA and RNA
2.3.2 Selfish Genes
2.4 Gene Duplication
3. Evolutionary Computation
3.1 Introduction to GA
3.2 Covariance Matrix Adaptation Evolution Strategy (CMA-ES)
3.3 Introduction to GP
3.4 Why GA and GP?
3.5 How to Pack a Knapsack?
3.6 GA Convergence
3.6.1 Wright-Fisher Model
3.6.2 Genetic Drift and Mutation Rate
3.6.3 Long Genotypes
3.6.4 Mutation Rate and GA Search
3.7 Introns and GA
3.7.1 How to Evolve a Bird?
3.7.2 Royal Road Function
3.7.3 Royal Road Function and Introns
3.7.3.1 Effectiveness of Introns
3.7.4 Introns in GP and Bloating
3.7.4.1 Introns in GP
3.7.5 Improvement of GP using Introns
3.7.5.1 Code Growth in GP
3.7.6 Why do GP Introns Emerge
3.7.7 Merits and Demerits of Introns
3.8 Estimation of Distribution Algorithm
3.9 Evolutionary Multi-objective Optimization: EMO
3.10 Interactive Evolutionary Computation (IEC)
3.11 Gene Duplication in GP
3.12 Selfish Genes: Revisited
4. Swarm Intelligence
4.1 Ant Colony Optimization (ACO)
4.1.1 Collective Behaviors of Ants
4.1.2 Simulating the Pheromone Trails of Ants
4.1.3 ACO using a Pheromone Trail Model
4.2 Particle Swarm Optimization (PSO)
4.2.1 Collective Behavior of Boids
4.2.2 PSO Algorithm
4.2.3 Comparison with GA
4.3 Firefly Algorithms
4.4 Cuckoo Search
4.5 Cat Swarm Optimization (CSO)
4.6 Swarms for Knapsack Problems
4.7 Swarm for Pareto-optimization
5. Deep Learning and Evolution
5.1 CNN and Feature Extraction
5.2 Autoencoders
5.3 Let us Fool the Neural Network
5.3.1 Generative Adversary Networks: GAN
5.3.2 Generating Fooling Images
5.4 LSTM
5.5 What is Neural Darwinism?
5.6 Neuroevolution
5.7 Let us Drive a Racing Car and Control a Helicopter
5.8 NEAT and HyperNEAT
5.9 CPPN and Pattern Generation
5.10 El Greco Test
6. Deep Swarms and Evolution
6.1 ACO for Construction of Evolutionary Trees
6.1.1 Phylogenetic Tree Derivation
6.1.2 Estimation using the Maximum Likelihood Method
6.1.3 How do Ants Search Trees?
6.1.4 ACO Simulation Results
6.2 Evolutionary Optimization Extended by Deep Learning
6.3 Preventing Overfitting of LSTMs using ACO
6.3.1 LSTM and Overfitting Problem
6.3.2 Optimizing the Structure of Neural Networks using ACO
6.3.3 ACO for LSTMs (ACOL)
6.3.4 Experiments with ACOL
6.3.5 Results of Ants
6.4 Deep Interactive Evolution
6.4.1 GAN and DeepIE
6.4.2 DeepIE3D
6.4.3 Deep Interactive Evolution Based on Graph Kernel: DeepIE3DGK
7. Emergence of Intelligence
7.1 Genes of Culture – Memes
7.2 Culture also Evolves
7.3 How the Brain is Created: Darwin among the Machines
A. Software Packages
A.1 Introduction
A.2 Multi-objective Optimization by GA
A.3 MindRender and MindRender/AIDrill
References
Indices